Mitochondrial

Acetyl-L-carnitine (ALCAR) is the acetyl ester of L-carnitine and the body's main carnitine ester. It is sold as a supplement and, in some countries, as a prescription drug. Meta-analyses find that it eases depressive symptoms, mostly in older adults and across small, inconsistent trials, and that it reduces diabetic nerve pain at doses above 1.5 g a day, on very low-certainty evidence. It did not slow Alzheimer's disease in two year-long trials, and it made chemotherapy nerve damage worse in a large one.

Alpha-lipoic acid (ALA) is a sulfur-containing mitochondrial cofactor, used as a racemic drug for the pain and pins-and-needles of diabetic nerve damage. Meta-analyses of randomized trials support it for diabetic neuropathy symptoms. Its effects on weight and blood sugar are real but small. Supplements come as the racemic mix, as pure R-ALA, or as Na-R-ALA, a sodium salt of R-ALA made to absorb better.

ATX-304, formerly O304, is a pan-AMPK activator from Betagenon developed as a type 2 diabetes drug and exercise mimetic. In a phase IIa trial in people with type 2 diabetes already on metformin, it lowered fasting glucose and insulin resistance and improved microvascular perfusion. In a 2026 phase 1b trial in 23 adults with obesity and prediabetes, 400 mg a day for 8 weeks was reported to lower liver fat, visceral fat and triglycerides and to raise resting metabolic rate, with minimal weight loss; so far that trial exists only as a poster abstract and a company press release. Newer mouse work extends it to fatty liver and kidney injury. The developer's patents describe a milled sodium salt: in volunteers, about a quarter of the dose in enteric capsules matched the blood levels of the original suspension, and rabbit work suggests both the salt and the milled capsule formulation contribute.

BAM15 is a mitochondrial uncoupler. It lets protons leak back into mitochondria so fuel is burned without making ATP, and unlike older uncouplers it does not depolarize the cell's outer membrane. In obese mice it reduced fat mass and improved insulin sensitivity without changing food intake, lean mass or body temperature. It has not been tested in humans.

Coenzyme Q10 (CoQ10) is part of the mitochondrial energy chain and one of the most widely used supplements. Its best evidence is in chronic heart failure. There, the Q-SYMBIO trial found fewer major cardiovascular events and deaths over two years. For statin muscle pain, migraine and blood pressure, the results are mixed or modest.

Elamipretide (SS-31) is a four-amino-acid peptide that concentrates in the inner mitochondrial membrane and binds cardiolipin, a membrane lipid that mitochondria need to form their inner folds (cristae). In September 2025 the US FDA gave it accelerated approval, as Forzinity, to improve muscle strength in Barth syndrome. Its randomized trials have mostly missed their primary endpoints, in Barth syndrome, mitochondrial myopathy, heart attack and macular degeneration. The positive signals come mainly from open-label extensions and secondary measures.

HDAP2 is a synthetic four-amino-acid peptide, biotin-D-Arg-Phe-Phe-D-Arg-amide, designed to bind cardiolipin, the signature lipid of the inner mitochondrial membrane, and hold it in its normal bilayer form. In mice it preserved retinal ganglion cells after optic nerve crush, and both the cells and their axons in an inherited glaucoma model, without lowering eye pressure. All of the evidence is from cells and mice, from the lab that invented and patented it. It has not been tested in people.

Humanin is a 24-amino-acid peptide encoded not in the nucleus but inside mitochondrial DNA, within the 16S rRNA gene. It was found in 2001 by screening a surviving region of an Alzheimer's brain for factors that protect neurons from amyloid-beta toxicity, which is an unusually direct origin story. It belongs to the same class as mots-c - mitochondria-derived peptides, or mitokines, which appear to signal cellular stress from mitochondria to the rest of the body. The preclinical literature is broad: cytoprotection against apoptosis, cardiovascular effects, roles in cancer and degenerative disease and in the testis. There is no human trial. Circulating humanin declines with age, which is interesting and is not the same as a reason to inject it.

Indolepropionamide (IPAM) is the amide of indole-3-propionic acid, a small indole related in structure to melatonin. Its inventors made it to get an antioxidant indole into the brain more easily, then reported finding a matching trace compound in rat brain. In their one published study it protected isolated rodent brain mitochondria at 10 nM and more than tripled the lifespan of a rotifer. That 2010 paper and the same authors' abandoned patent application are the whole evidence base: no other laboratory has published an experiment with IPAM, and it has never been tested in a mammalian disease model or in people.

J147 is a synthetic small molecule from the Salk Institute. It descends from curcumin through a series of hybrid compounds and was chosen because it protected neurons in six different cell models of brain aging. Fed to mice, it improved memory in normal animals, in Alzheimer's-model mice treated late in the disease and in rapidly aging SAMP8 mice. Its target turned out to be the mitochondrial ATP synthase, which it partly inhibits, switching on AMPK. A phase 1 trial in healthy volunteers finished in 2020 and has never reported results, so there is no published human data at any dose.

Mitochonic Acid 5 (MA-5) is a synthetic indole compound studied for mitochondrial disease. It binds mitofilin/MIC60, a protein involved in organizing the inner mitochondrial membrane. Human development has reached phase I safety testing; clinical benefit remains unproven.

MOTS-c is a 16-amino-acid peptide encoded in mitochondrial DNA, not in the cell nucleus. In mice, injections prevented diet-induced obesity and insulin resistance and improved running capacity at every age tested. In people, exercise raises the body's own MOTS-c, but no trial has given MOTS-c to humans, and measurements of it in blood disagree between methods.

Myricetin is a flavonol found in berries, vegetables, tea and wine. It is studied for glucose regulation, mitochondrial function and neuroprotection, but the evidence for the isolated molecule remains preclinical. Small human trials have used blends containing myricetin; none separates its contribution from the other ingredients.

NAD+ (nicotinamide adenine dinucleotide) is a coenzyme in every cell. It carries electrons in energy metabolism and is used up by sirtuins, PARPs and CD38. Clinics sell it as an intravenous drip, but no trial has tested NAD+ infusions for any health outcome. In a small study, infused NAD+ disappeared from the blood for the first 2 hours. Most human data come from its oral precursors NR and NMN. They reliably raise NAD+, but their effects on health outcomes are mixed and often null.

PQQ (pyrroloquinoline quinone) is a redox cofactor first found in bacteria. Mammals do not make it but take in trace amounts from food. In cells it switches on mitochondrial biogenesis through the PGC-1α pathway. In humans, small 12-week trials of PQQ disodium salt reported better memory and attention scores.

SANA is a nitroalkene derivative of salicylate, the parent of aspirin, developed for obesity and type 2 diabetes. In obese mice it reduced weight, liver fat and insulin resistance by raising creatine-dependent energy expenditure in fat tissue. MVD1 is the name its first human trial ran under. That trial was a small, 15-day phase 1 study, which reported good tolerability and early effects on body weight and glucose.

SLU-PP-332 is a synthetic agonist of all three estrogen-related receptors (ERRα, β and γ), strongest at ERRα, and the best-known "exercise mimetic" research compound. In mice it switched on an aerobic exercise gene program, increased oxidative muscle fibers and improved endurance. In obese mice it raised energy expenditure and fat oxidation, reduced fat mass and improved glucose tolerance. It has never been tested in humans.

SLU-PP-915 is a pan-ERR agonist from the lab that made slu-pp-332. It is a chemically distinct thiophene boronic acid and, unlike SLU-PP-332, it is orally bioavailable. In mice it improved aerobic exercise performance as much as SLU-PP-332 did and protected the heart in a heart-failure model. It has never been tested in humans.

Ubiquinol is the reduced, electron-carrying form of coenzyme-q10, sold as a better-absorbed version. One small study found it raised blood CoQ10 more than ubiquinone. A randomized crossover study in older adults found no significant difference. Its most notable trial is in multiple-system atrophy, where high doses improved a motor score.

Urolithin A is not something plants make. It is what some people's gut bacteria make out of the ellagitannins in pomegranates, walnuts and berries, and only about one person in seven produces enough of it to show up in blood without a supplement. Synthetic urolithin A was identified as the first natural compound to induce mitophagy — the pathway that clears worn-out mitochondria. Four randomized trials have now tested it in people. The two muscle trials both missed their primary endpoint and both reported gains on secondary measures; a 2025 trial found changes in T-cell composition and metabolism; and a trial in trained runners found no performance benefit.